[SPARK-19337] [ML] [Doc] Documentation and examples for LinearSVC #16968
Conversation
|
Test build #73020 has finished for PR 16968 at commit
|
|
I see. I will drop the R example here, whichever PR goes in later can finish the document update. |
docs/ml-classification-regression.md
Outdated
| regression, or other tasks. Intuitively, a good separation is achieved by the hyperplane that has | ||
| the largest distance to the nearest training-data point of any class (so-called functional margin), | ||
| since in general the larger the margin the lower the generalization error of the classifier. LinearSVC | ||
| in Spark ML supports binary calssification with linear SVM. Internally, it optimizes the |
There was a problem hiding this comment.
calssification -> classification
docs/ml-classification-regression.md
Outdated
| A [support vector machine](https://en.wikipedia.org/wiki/Support_vector_machine) constructs a hyperplane | ||
| or set of hyperplanes in a high- or infinite-dimensional space, which can be used for classification, | ||
| regression, or other tasks. Intuitively, a good separation is achieved by the hyperplane that has | ||
| the largest distance to the nearest training-data point of any class (so-called functional margin), |
There was a problem hiding this comment.
"largest distance" -> "longest distance"? I think?
There was a problem hiding this comment.
Thanks for the comment. I think both large and long can be used to describe distance, wherever large is more suitable to describe the numeric margin. Please let me know if you have a strong preference.
|
title should say |
hhbyyh
changed the title
|
Thanks for the comment @felixcheung |
|
Test build #73071 has finished for PR 16968 at commit
|
felixcheung
left a comment
felixcheung
left a comment
There was a problem hiding this comment.
LGTM.
I'll leave this out for a day in case anyone else can comment.
docs/ml-classification-regression.md
Outdated
| regression, or other tasks. Intuitively, a good separation is achieved by the hyperplane that has | ||
| the largest distance to the nearest training-data points of any class (so-called functional margin), | ||
| since in general the larger the margin the lower the generalization error of the classifier. LinearSVC | ||
| in Spark ML supports binomial classification with linear SVM. Internally, it optimizes the |
There was a problem hiding this comment.
actually, is there a reason you change this to binomial classification?
There was a problem hiding this comment.
just to be consistent with LR. But I'm not sure if it's the common expression.
There was a problem hiding this comment.
do you have a link? I think binary classification is more commonly used
There was a problem hiding this comment.
FWIW I have never head the term binomial classification and it doesn't show up in a Google search. I think it was a typo.
felixcheung
left a comment
felixcheung
left a comment
There was a problem hiding this comment.
let's change binomial classification
|
Thanks for the review. Updated to binary. |
|
Test build #73132 has finished for PR 16968 at commit
|
|
merged to master. |
## What changes were proposed in this pull request? Documentation and examples (Java, scala, python, R) for LinearSVC ## How was this patch tested? local doc generation Author: Yuhao Yang <[email protected]> Closes apache#16968 from hhbyyh/mlsvmdoc.
6 participants
What changes were proposed in this pull request?
Documentation and examples (Java, scala, python, R) for LinearSVC
How was this patch tested?
local doc generation